Media intelligence

The work I have done as as guest professor at Södertörn university during autumn 2015, within a project run by my colleague Lars Degerstedt at the School of Natural Science, Technology and Environmental Studies, has come to an end, but is now slowly developing into a jointly written article. Below is the framework around the concept of “media intelligence” that we are trying to develop. The article ought to be finished within a week or so.

As the arguably leading business intelligence company in Scandinavia, Cision—with a history dating back to 1892 under the name, Svenska telegrambyrån, a company that provided press clipping services in Sweden—boasts online of being a global enterprise within communication and media intelligence. “Cision serves the complete workflow of today’s communications, social media and content marketing professionals” (Cision, 2016). Yet, what does the specific notion of media intelligence actually mean? Basically, it refers to various forms of updated media monitoring practices—both manual and automatic—foremost regarding print and broadcast media. Obviously, online media has also played an increasingly important role for the media intelligence business during the last two decades. Media intelligence can also be understood by comparing it to adjacent intelligence fields. There are for example analogies to the notion of intelligence operations, as these are executed within the military domain. Military intelligence is, in short, a defense discipline that exploits a number of information collection practices and strategies in order to provide guidance for commanders. Beside military usages, computational intelligence also offers some insights. In automatised versions, crawling the social web to identify relevant conversations for example, media intelligence resembles artificial intelligence, thus linking the notion of intelligence to both machines and systems as well as networks. Rather than sticking to a strict scholarly media and communication research perspective, media intelligence should hence be understood within a larger framework of military and artificial intelligence, as well as more traditional forms of business and market analytics, adjacent to fields as strategic communication, public relations and communication management.

If military intelligence is sometimes divided into strategic, operational and tactical intelligence, the same basically goes for media intelligence. Within the commercial business sector media intelligence is often roughly divided into: (A.) business intelligence (on a particular company level), and (B.) competitive intelligence (between similar companies regarding for example shared markets). In general, the latter is different from the former since it uses and analyses data outside company firewalls. However, during the last decade—mainly due to profound technological changes brought about by digitisation—the specificities of and boundaries between business intelligence and competitive intelligence have been modified. When society is gradually turning into a market of different mediated “value networks”, as Sven Hamrefors argued in 2010, “communication functions can no longer stay in their restricted domains and only deal with traditional communication issues” (Hamrefors, 2010). On the one hand, intelligence on a strategic business-to-business-level hence should not be separated from a more practical business-to-consumer-level. On the other hand, an increasing number of companies (foremost within the tech domain) operate in different market segments, making it more or less impossible to intelligence and monitor all relevant markets. The most obvious example is Google, which started with search, soon began making operating systems, run different forms of content platforms—and now even produces cars. The same can be said of an international media group as Schibsted who does business in a number of different digital domains (publishing, online marketplaces and services).

In more concrete terms, media intelligence uses data and computer science methods to analyse both social media and editorial media content. In general, within business intelligence today, it is often argued that a brave new world of insight awaits intelligence companies and their customers, if they have the courage—or the financial abilities—to analytically start working with the exponentially growing volumes of unstructured and semi-structured data—especially from new data sources as machines, sensors, logs, and (non-textual) social media and streaming data. Basic implementation for media intelligence involves curating data, keyword references and semantic analyses, as well as natural language processing via machine learning algorithms. In essence, most practices and operations are concerned with turning text into data for analysis. Text is hence still the dominant modality for most media intelligence operations, yet other modalities (images, sound and video) have during the last decade become increasingly important, especially in different social forms.

Most forms of media intelligence, departs from the ways in which we currently are enmeshed in an “interconnected communications ecosystem wherein social and traditional media sources feed each other for stories and conversation, and those conversations are supercharged by social technology” (Nuccio, 2015). Media intelligence hence refers to computational solutions that tries to synthesise innumerable online conversations into (more or less) appropriate insights that allow companies and organisations to manage, and sometimes even measure content performance and trends—with the paramount purpose to better forecast business strategies. As is to be expected, companies working within the intelligence business sector offer quite different suggestions as to what media intelligence actually means: “Media intelligence is the process of gathering all the data available through social media and news media outlets and analyzing the data to allow for better business decision making”, according to CustomScoop (2016); Volicon is said to be the “leading provider of enterprise media intelligence solutions serving the needs of broadcasters, networks, cable operators, and governments worldwide” (Volicon, 2016); and M-Brain states that its media intelligence solutions are designed to “monitor and measure your publicity and reputation” (M-Brain, 2016). What these companies have in common, however, is that they gather massive amounts of data points from user-generated content on social media sites, blogs and comment fields, combining these with traditional mass media output and other forms of publicly open data, all in order to provide—and ultimately sell—real-time insights and suggestions based on verifiable data. Media intelligence is hence, always about selling trust.

Some scholars have argued that during recent years media intelligence has witnessed a social transition, from various forms of social media monitoring to computational driven (social) media intelligence. The latter is said to be better equipped to minister noisy sociality, with an ability to uncover valuable insights hidden in the social media chatter (Moe & Schweidel, 2014). Research related to various forms of military intelligence, have furthermore tried to identify and forecast civil unrest and radical mobilization by mining textual content in open-source social media (Agarwal & Sureka, 2015). In general, social media intelligence is based on a rudimentary data management model where social data is segmented—from automatically categorised subsets of social data, to customising rules or filtering, based on criteria like date, location, web page type, sentiment and gender. Segmentation can also be done on more specific data, for example regarding Twitter or Facebook statistics (retweets, likes, comments, media type etcetera.) In essence, data management within social media intelligence collects massive volumes of data and separates it into structured and manageable packages that can help answer particular questions via different forms of machine learning and data mining. The notions are similar; machine learning and data mining often overlap. But they also differ in that machine learning focuses on prediction—based on known properties within the collected data—whereas data mining is about the discovery of unknown properties in the same data.

Then again, if sociality has recast media intelligence during the last decade, modalities of media content is another alteration. Traditionally, business intelligence companies have relied on textual offerings—basically because machine learning algorithms need text based documents (and databases) to be able to perform automatic analyses of large scale data sets. The modality of text has, in short, been default, not the least apparent in the ways companies within the business intelligence sector have advertised themselves: “Keep track of what is written about you, your company or your competitors” (Cision, 2016); “Infomap:r is a system for predictive analytics and text mining” (Infomap:r 2016); “We keep your organization up to date on what is written and said about you and your business environment” (Newsmachine, 2016).

We argue however, that if online media has experiencing a shift towards the social, at the same time online interaction has increasingly been enriched with images, sound and videos. These new media modalities have brought forth changes that are currently having profound effects on the media intelligence business. The before mentioned infographic from Domo serves as a vivid illustration of challenges in both social and non textual media form(at)s facing the intelligence business. If YouTube has been the epitome of an ever increasing non textuality of the information landscape during the last decade (Snickars & Vonderau, 2009), the blended mix of Facebook posts in different modalities in many ways acts as its social counterpart. However, ‘social’ and ‘multimedia’ is also converging. In January 2014, for example, Facebook announced an increasing shift towards visual content, “especially with video … In just one year, the number of video posts per person has increased 75% globally and 94% in the US” (Facebook, 2015a). In fact, during 2014 Facebook had an average of more than one billion video views every day. Social video is thus an increasing trend, and the release of Facebook Instant Articles in May 2015 was consequently aimed towards the ability of watching audiovisual news material seamlessly. “Zoom in and explore high-resolution photos by tilting your phone. Watch auto-play videos come alive as you scroll through stories. Explore interactive maps, listen to audio captions” (Facebook, 2015b).

If business intelligence in automated forms have traditionally relied on text mining to monitor, detect and analyse plain online text sources, the transition to new social media modalities hence causes difficulties, both conceptually and technologically. If humans can perceive their surroundings naturally in visual form, “this undertaking is quite challenging for machines”, according to Damian Borth. Within the field of computer science machine learning in non-textual forms is, in short, utterly complicated. “The lack of correspondence between the low-level features that machines can extract from videos (i.e., the raw pixel values) and the high-level conceptual interpretation a human associates with perceived visual content is referred to as the semantic gap” (Borth, 2014). Nevertheless, even if media intelligence is struggling with new social media modalities, companies within business intelligence are also increasingly trying to cope with this semantic gap. The media intelligence company Opoint for example, are said to be “the only player in the market that monitors real-time radio and TV… [and analyses] all types of media.” Sound bites are, for example, delivered direct to customers; that is, Opoint are not using speech recognition software transforming sound into text (Opoint, 2016). Another similar company, Lissly asserts that its “tool collects, sorts and visualizes data from different digital media.” Lissly offers its customers to be able to “listen to the conversations in your market” (Lissly, 2016). In fact, the metaphor of listening is often used today within the media intelligence business, a semantic indication that other media modalities than text are becoming continuously more important: “Notified takes social media listening and management to the next level” (Notified, 2016); “We believe that the world will be a little bit better if we listen more” (NewsMachine, 2016); “There is power in listening to what customers and stakeholders are saying about your business through social media platforms” (M-Brain, 2016).

The major challenge that all business intelligence (based on other media modalities than text) are faced with today, is the somewhat paradoxical movement away from the content of communication towards the medium of communication. On the one hand, there needs to be a market demand for this kind of transition to occur—that is: the request of monitoring other (or new forms of digital) media (as data streams). At present such demand is still by and large insufficient, yet mainly because media intelligence algorithms still cannot produce appropriate results from non textual information. On the other hand, the transition—or perhaps dialectics between content and medium—also resonates in an interesting way with debates within classical media theory as to what constitutes the bias of communication. Content and medium have, in short, always been intertwined. Following Harold Innis in the 1950s, and his belief that the stability of cultures depended on the balance and proportion of each particular media form—from clay to papyrus—he claimed that each medium embodied a certain bias in terms of organisation and control of information (Innis, 1950). Marshall Mcluhan’s 1960s media theory, where the medium itself constituted the message, followed Innis ideas closely—or as McLuhan famously stated: “I am pleased to think of my own book Gutenberg Galaxy as a footnote to the observations of Innis” (McLuhan, 1964). Then again, similar ideas have also been put forward within research fields associated with media intelligence. So called “media richness theory”, for example, have been developed within organisation and management studies to describe a medium’s ability to reproduce information. Basically, communications that require a long time to enable understanding are lower in richness. “Rich media are personal and involve face-to-face contact between managers, while media of lower richness are impersonal and rely on rules, forms, procedures, or data bases”, according to Richard L. Daft and Robert H. Lengel (1986). Following media richness theory, face-to-face communication is thus perceived as the richest media form since it provides immediate feedback, and such ideas also resonates in the contemporary work of Sherry Turkle, as in her new book, Reclaiming Conversation. The Power of Talk in a Digital Age (2015). Daft and Lengel’s media richness theory was introduced in the 1980s to help organisations cope with various forms of communication challenges. In an equivalent way these ideas can be today be helpful when trying to explain the conceptual and technical obstacles facing the media intelligence business, regarding for example social competitive intelligence and media analytics.